The primary purpose of the proposed study is to conduct a rigorous process and outcome evaluation of a community-wide health service prevention/intervention program to prevent drunk driving in Frederick, Maryland. The intervention will include two central components, each of which has been suggested in related literature to contribute towards reducing rates of drinking and driving and related traffic accidents and injuries, namely Designated Driver Programs (DDPs) and providing 'Safe Rides' home (SRs) for drunk drivers. This study will be the first to carefully investigate the combined impact of these two programs. After an extensive baseline assessment on drinking and driving attitudes and related risk behavior in Frederick and in a control community, the two community-based interventions will be introduced in the 'experimental' community, Frederick, Maryland. Using a wide range of publicity sources and techniques, a DDP will be implemented to promote the use of Designated Drivers (DDs) by patrons of drinking establishments, and by guests at private gatherings where alcohol is served. DDs in participating establishments will receive various incentives for participating as a means of recognition for being responsible companions for drinkers. A second large-scale effort will be implemented simultaneously that will offer drinkers, who are intoxicated and perceive no other options to drinking and driving, with a complimentary ride home (from bars, restaurants that serve alcohol, or from private parties). In exchange for completing a study questionnaire, these individuals will be provided a complimentary ride to their cars the next day, or if not, will receive a standard incentive. In addition to studying DDs and SR service customers, an extensive assessment of community members' drinking and driving attitudes and behavior, and barroom patrons' opinions and behavior, will be conducted yearly at an experimental and control site throughout the 4-year grant period. Analyses will examine the impact of the publicity and prevention/intervention effort on (1) improving the attitudes and drinking and driving behavior of drinking establishment patrons, people who use the DDP and/or SR services, and community members at large; and on (2) reducing the incidence of drinking and driving arrests and crashes in the targeted community. The study design will permit us to examine differences in impacts between (1) people differentially exposed to program publicity in the target community, and on (2) members of the target versus the control community that will not receive these services. If this relatively inexpensive prevention/intervention package has its anticipated effect on drinking and driving attitudes and behavior in Frederick, it could be readily implemented in other communities. Maryland State officials have expressed a specific interest in replicating the program should it prove effective.